Assistive devices have been developed to help those with tetraplegia better interact with their environment. Complex assistive devices, such as neuroprostheses or robotics, require users to generate many commands to control the multidimensional movement of these devices. However, individuals with severe tetraplegia are limited to generating commands from the neck and above. Accessing movement-intent from the brain will increase the number of command signals available for control of these devices. Actual and imagined movements of specific body parts generate changes in field potentials over brain areas associated with those body parts. These movement-related changes can be recorded from the scalp, the brain surface, or anywhere in between. Field potential changes can be used to control multiple aspects of an assistive device. However, the farther the recording electrodes are from the brain the more difficult it is to distinguish brain activity generated by movement-intent of adjacent body parts. The goal of this study is to quantify how reliably the intended-movement of adjacent body parts can be differentiated using field potentials recorded at the scalp and under the scalp. Individuals with high-level spinal cord injury will undergo scalp and subdermal field potential recording initially using standard EEG electrode spacing. The ability to differentiate brain signals generated by adjacent body parts will be quantified. Then, customized electrode placements will be explored to further increase the ability to distinguish between body parts and thereby increase the potential number of command signals available to users of assistive devices. In addition, the ability for an individual with a high-level spinal cord injury to command a virtual upper-extremity neuroprosthesis in real-time using scalp or subdermal EEGs recorded from optimal electrode locations will be evaluated. Determining the optimal electrode locations will help guide the design and surgical placement of subdermal and skull-embedded electrodes where the spatial resolution must be estimated beforehand. Optimizing the spatial resolution and assessing the utility of field potential recordings will help improve permanently implantable command systems for individuals with tetraplegia;increasing their ability to interact with their environment using assistive devices.